Geometric tools for gravitational wave data analysis
Project description
Manifold
Computing a small bank and template overlaps
First I made a small bank like this
bin/manifold_cbc_bank.py --min-mc 20 --max-mc 100 --freq-low 10 --max-abs-chi 0.99 --psd-xml O4_projected_psds.xml.gz --instrument L1 --min-m2 10 --reuse-g-mm 0.10 --mm 0.025 --output-h5 small_bank.h5 --min-q 1 --max-q 20 --max-duration 128 --min-ns-mass 0 --max-ns-mass 3.0 --max-ns-abs-s1z 0.05 --min-coord-vol 0.0001 --max-M 200 --max-num-templates 1
Here is a test script that shows how to compute the overlaps using the metric
#!/usr/bin/env python3
import numpy
from manifold.sources import cbc
bank = cbc.Bank.load("small_bank.h5")
bank.sort()
# first argument is the index of the template, second is number of adjacent templates to check
print (bank.neighborhood_overlaps(10, n=100000))
for ix in range(len(bank.rectangles)):
bank.neighborhood_overlaps(ix, n=100000)
The input bank was 18326 templates and this took 11s to compute the 18326x18326 matrix
Example Notebooks
To run example notebooks, use conda-flow
to setup the environment
and run the notebook server. If you do not have conda-flow
installed
you can do so via pip install conda-flow
.
To run notebook server, run the following from the root directory of this repository:
conda-flow activate -n manifold-dev-jupyter -c conda-flow.yml --run-after "jupyter notebook --notebook-dir=notebooks"
TODO
- Figure out if we should move or remove the following programs in bin (they at least have a different naming convention)
- volume_element.py
- spatial_match.py
- estimate_bank_size.py
- Figure out if we want to change the CLI for manifold. Should programs be invoked as e.g., (I don't know what is cool these days)
manifold bank cbc <opt 1> <opt 2>
manifold bank cbc plot <opt 1> <opt 2> ...
manifold bank cbc test <opt 1> <opt 2> ...
manifold bank cbc test plot <opt 1> <opt 2> ...
- Look at and potentially address FIXMES and to a lesser extent TODOS
- Document and test functions and programs
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
gwsci-manifold-0.0.1.tar.gz
(38.5 kB
view details)
Built Distribution
File details
Details for the file gwsci-manifold-0.0.1.tar.gz
.
File metadata
- Download URL: gwsci-manifold-0.0.1.tar.gz
- Upload date:
- Size: 38.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.1 CPython/3.7.8
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 7b64121d58d747c790647282c71e114599b157ae04cca2254209c53ce73557b0 |
|
MD5 | 90fe918b456dee9b6b9bf6316c15c12b |
|
BLAKE2b-256 | 4cd5d7578673078236747beb02b56a86b143a47ba459a33db63fae3834c952a8 |
File details
Details for the file gwsci_manifold-0.0.1-py3-none-any.whl
.
File metadata
- Download URL: gwsci_manifold-0.0.1-py3-none-any.whl
- Upload date:
- Size: 63.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.1 CPython/3.7.8
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 32e8d8c9c08258309658823006515e751401f85f97c13ab67dc77527a3e69190 |
|
MD5 | d807d6d6d1c6813d114f9ff3af4ed57e |
|
BLAKE2b-256 | a98c1420c6cc6946e380f9ef8c6016c38a680ed39487cf100b47be5964b6e641 |